ca-aird/airtraffic2019

---
task_categories:
- graph-ml
---
**Description.** The Bureau of Transportation Statistics, under the United States Department of Transportation, monitors and reports on the on-time performance of domestic flights operated by major airlines.
The datasets for [2019](https://www.kaggle.com/datasets/flight-delay-and-causes.) and [2015](https://www.kaggle.com/datasets/usdot/flight-delays.) are publicly available to enable analyses of flight delays and airport performance.
Each dataset consists of flight records for to the corresponding year.
The flight records contain information on source and destination airports and scheduled and actual departure and arrival times.
To investigate the impact of weather conditions on flight delays, we have supplemented the flight datasets with weather data from [Open-Meteo](https://open-meteo.com/en/docs), an open-source weather API [Zippenfenig, 2023].
We extracted weather conditions at the scheduled departure times for both origin and destination airports, enabling the analysis of weather-related delay predictions.
This integration enhances the datasets’ capabilities, allowing for a more comprehensive understanding of weather’s role in flight delay dynamics.

**Graph Construction.** We construct a graph where airports are represented as vertices and flights as edges.
The vertices lack feature attributes. Each edge, symbolizing a flight, is associated with a timestamp indicating the flight date and a feature vector derived from weather conditions at the endpoint airports, comprising: daily precipitation sum, maximum and minimum daily air temperature, maximum wind speed and gusts on the flight day.
Edge targets are defined as arrival delay normalized by flight duration, representing the delay outcome.

原始信息汇总

数据集概述

数据集来源

• 由美国交通运输部下属的交通统计局提供。

数据集内容

• 包含2019年和2015年的国内航班记录数据，每个数据集包含该年度的航班记录。
• 航班记录信息包括：
  • 出发和到达机场
  • 计划和实际的起飞及到达时间
• 为了分析天气条件对航班延误的影响，数据集整合了来自Open-Meteo的天气数据，包括：
  • 每日降水量总和
  • 日最高和最低气温
  • 最大风速和阵风

数据集用途

• 用于分析航班延误和机场性能。
• 通过整合天气数据，增强了对天气在航班延误中作用的全面理解。

数据集结构

• 图结构：
  • 机场作为顶点，航班作为边。
  • 顶点无特征属性。
  • 边包含时间戳（航班日期）和基于端点机场天气条件的特征向量。
  • 边的目标定义为到达延误与航班持续时间的比值，代表延误结果。